Preliminary communication

https://doi.org/10.17559/TV-20180305095253

Application of SVM Models for Classification of Welded Joints

Dejan Marić orcid.org/0000-0002-0142-1750 ; Mechanical Engineering Faculty in Slavonski Brod, Josip Juraj Strossmayer University of Osijek, Trg Ivane Brlić Mažuranić 2, HR-35000 Slavonski Brod, Republic of Croatia
Miroslav Duspara orcid.org/0000-0003-4249-9279 ; Mechanical Engineering Faculty in Slavonski Brod, Josip Juraj Strossmayer University of Osijek, Trg Ivane Brlić Mažuranić 2, HR-35000 Slavonski Brod, Republic of Croatia
Tomislav Šolić ; Mechanical Engineering Faculty in Slavonski Brod, Josip Juraj Strossmayer University of Osijek, Trg Ivane Brlić Mažuranić 2, HR-35000 Slavonski Brod, Republic of Croatia
Ivan Samardžić ; Mechanical Engineering Faculty in Slavonski Brod, Josip Juraj Strossmayer University of Osijek, Trg Ivane Brlić Mažuranić 2, HR-35000 Slavonski Brod, Republic of Croatia

Abstract

Classification algorithm based on the support vector method (SVM) was used in this paper to classify welded joints in two categories, one being good (+1) and the other bad (−1) welded joints. The main aim was to classify welded joints by using recorded sound signals obtained within the MAG welding process, to apply appropriate preprocessing methods (filtering, processing) and then to analyze them by the SVM. This paper proves that machine learning, in this specific case of the support vector methods (SVM) with appropriate input conditions, can be efficiently applied in assessment, i.e. in classification of welded joints, as in this case, in two categories. The basic mathematical structure of the machine learning algorithm is presented by means of the support vector method.

Keywords

Classification; Machine learning; Sound signal; SVM model; Welding

Hrčak ID:

219547

URI

https://hrcak.srce.hr/219547

Publication date:

24.4.2019.

Visits: 1.518 *